qkeras QDense_batchnorm?

QDense_batchnorm?

Open thesps opened this issue 3 years ago • 2 comments

I see you have qconv2d_batchnorm layer which folds the weights of the two layers and then quantizes.

We're bringing support for that to hls4ml, and it should help us save some resources & latency.

I'm wondering, do you plan to add the equivalent combined QDense + BatchNormalization layer to QKeras?

Mar 15 '21 15:03 thesps

@thesps great and glad to see this helps! Yes, QDenseBatchnorm is one of our TODO items, but we have other higher priority tasks at this moment.

Mar 15 '21 16:03 zhuangh

https://github.com/google/qkeras/pull/74

Jun 17 '21 04:06 zhuangh